[dm-devel] [RFC PATCH] fstests: Check if a fs can survive random (emulated) power loss
Qu Wenruo
quwenruo.btrfs at gmx.com
Thu Mar 1 09:27:23 UTC 2018
On 2018年03月01日 16:39, Amir Goldstein wrote:
> On Thu, Mar 1, 2018 at 7:38 AM, Qu Wenruo <wqu at suse.com> wrote:
>> This test case is originally designed to expose unexpected corruption
>> for btrfs, where there are several reports about btrfs serious metadata
>> corruption after power loss.
>>
>> The test case itself will trigger heavy fsstress for the fs, and use
>> dm-flakey to emulate power loss by dropping all later writes.
>
> So you are re-posting the test with dm-flakey or converting it to
> dm-log-writes??
Sorry, I just noticed the date right now.
I was formatting the patch to the wrong place with old patches.
Please ignore this patch.
Thanks,
Qu
>
>>
>> For btrfs, it should be completely fine, as long as superblock write
>> (FUA write) finishes atomically, since with metadata CoW, superblock
>> either points to old trees or new tress, the fs should be as atomic as
>> superblock.
>>
>> For journal based filesystems, each metadata update should be journaled,
>> so metadata operation is as atomic as journal updates.
>>
>> It does show that XFS is doing the best work among the tested
>> filesystems (Btrfs, XFS, ext4), no kernel nor xfs_repair problem at all.
>>
>> For btrfs, although btrfs check doesn't report any problem, kernel
>> reports some data checksum error, which is a little unexpected as data
>> is CoWed by default, which should be as atomic as superblock.
>> (Unfortunately, still not the exact problem I'm chasing for)
>>
>> For EXT4, kernel is fine, but later e2fsck reports problem, which may
>> indicates there is still something to be improved.
>>
>> Signed-off-by: Qu Wenruo <wqu at suse.com>
>> ---
>> tests/generic/479 | 109 ++++++++++++++++++++++++++++++++++++++++++++++++++
>> tests/generic/479.out | 2 +
>> tests/generic/group | 1 +
>> 3 files changed, 112 insertions(+)
>> create mode 100755 tests/generic/479
>> create mode 100644 tests/generic/479.out
>>
>> diff --git a/tests/generic/479 b/tests/generic/479
>> new file mode 100755
>> index 00000000..ab530231
>> --- /dev/null
>> +++ b/tests/generic/479
>> @@ -0,0 +1,109 @@
>> +#! /bin/bash
>> +# FS QA Test 479
>> +#
>> +# Test if a filesystem can survive emulated powerloss.
>> +#
>> +# No matter what the solution a filesystem uses (journal or CoW),
>> +# it should survive unexpected powerloss, without major metadata
>> +# corruption.
>> +#
>> +#-----------------------------------------------------------------------
>> +# Copyright (c) 2018 SuSE. All Rights Reserved.
>> +#
>> +# This program is free software; you can redistribute it and/or
>> +# modify it under the terms of the GNU General Public License as
>> +# published by the Free Software Foundation.
>> +#
>> +# This program is distributed in the hope that it would be useful,
>> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
>> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
>> +# GNU General Public License for more details.
>> +#
>> +# You should have received a copy of the GNU General Public License
>> +# along with this program; if not, write the Free Software Foundation,
>> +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
>> +#-----------------------------------------------------------------------
>> +#
>> +
>> +seq=`basename $0`
>> +seqres=$RESULT_DIR/$seq
>> +echo "QA output created by $seq"
>> +
>> +here=`pwd`
>> +tmp=/tmp/$$
>> +status=1 # failure is the default!
>> +trap "_cleanup; exit \$status" 0 1 2 3 15
>> +
>> +_cleanup()
>> +{
>> + ps -e | grep fsstress > /dev/null 2>&1
>> + while [ $? -eq 0 ]; do
>> + $KILLALL_PROG -KILL fsstress > /dev/null 2>&1
>> + wait > /dev/null 2>&1
>> + ps -e | grep fsstress > /dev/null 2>&1
>> + done
>> + _unmount_flakey &> /dev/null
>> + _cleanup_flakey
>> + cd /
>> + rm -f $tmp.*
>> +}
>> +
>> +# get standard environment, filters and checks
>> +. ./common/rc
>> +. ./common/filter
>> +. ./common/dmflakey
>> +
>> +# remove previous $seqres.full before test
>> +rm -f $seqres.full
>> +
>> +# real QA test starts here
>> +
>> +# Modify as appropriate.
>> +_supported_fs generic
>> +_supported_os Linux
>> +_require_scratch
>> +_require_dm_target flakey
>> +_require_command "$KILLALL_PROG" "killall"
>> +
>> +runtime=$(($TIME_FACTOR * 15))
>> +loops=$(($LOAD_FACTOR * 4))
>> +
>> +for i in $(seq -w $loops); do
>> + echo "=== Loop $i: $(date) ===" >> $seqres.full
>> +
>> + _scratch_mkfs >/dev/null 2>&1
>> + _init_flakey
>> + _mount_flakey
>> +
>> + ($FSSTRESS_PROG $FSSTRESS_AVOID -w -d $SCRATCH_MNT -n 1000000 \
>> + -p 100 >> $seqres.full &) > /dev/null 2>&1
>> +
>> + sleep $runtime
>> +
>> + # Here we only want to drop all write, don't need to umount the fs
>> + _load_flakey_table $FLAKEY_DROP_WRITES
>> +
>> + ps -e | grep fsstress > /dev/null 2>&1
>> + while [ $? -eq 0 ]; do
>> + $KILLALL_PROG -KILL fsstress > /dev/null 2>&1
>> + wait > /dev/null 2>&1
>> + ps -e | grep fsstress > /dev/null 2>&1
>> + done
>> +
>> + _unmount_flakey
>> + _cleanup_flakey
>> +
>> + # Mount the fs to do proper log replay for journal based fs
>> + # so later check won't report annoying dirty log and only
>> + # report real problem.
>> + _scratch_mount
>> + _scratch_unmount
>> +
>> + _check_scratch_fs
>> +done
>> +
>> +echo "Silence is golden"
>> +
>> +# success, all done
>> +status=0
>> +exit
>> diff --git a/tests/generic/479.out b/tests/generic/479.out
>> new file mode 100644
>> index 00000000..290f18b3
>> --- /dev/null
>> +++ b/tests/generic/479.out
>> @@ -0,0 +1,2 @@
>> +QA output created by 479
>> +Silence is golden
>> diff --git a/tests/generic/group b/tests/generic/group
>> index 1e808865..5ce3db1d 100644
>> --- a/tests/generic/group
>> +++ b/tests/generic/group
>> @@ -481,3 +481,4 @@
>> 476 auto rw
>> 477 auto quick exportfs
>> 478 auto quick
>> +479 auto
>
> + stress
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 520 bytes
Desc: OpenPGP digital signature
URL: <http://listman.redhat.com/archives/dm-devel/attachments/20180301/a1f094e3/attachment.sig>
More information about the dm-devel
mailing list